Attempted Hydrogenolysis of Amines with ... - ACS Publications

Infrared absorptions characteristic to the di- ethylaminotin grouping, Sn—N(C2H5)2, were ob- served at 1290 (m), 1173 (vs), 1010 (s), 880 (s), and 7...
2 downloads 0 Views 254KB Size
+

2Rk3nN(C2H5)2 C o n

+ H20+

+ 2(C2H6),NH (3)

(Rdh)?COs

( R = GH6, n-CaH7, or n-GHg)

Attempted Hydrogenolysis of Amines with Triphenyltin Hydride’ BLFRED STERS’ A N D

Infrared absorptions characteristic to the diethylaminotin grouping, Sn--N (C2H&, were observed a t 1290 (m), 1173 (vs), 1010 (s), 880 (s), and 780 (s) cm.-l, all being common to the three trialkyl(diethylamin0) tins herein described. These bands completely disappeared on exposure of the sample to air for several minutes on a sodium chloride plate, when characteristic bands of organotin carbonate appeared at 1560-1520 (vs), 1370 (vs), 1070 (m), and 840 (m) cm.-’ instead. These results show that the organotin-nitrogen bonding is rather stable to heating, as indicated by no change upon distillation, but is highly reactive toward proton-containing reagents. Experimental

Analyses .-A microelementary analysis for carbon and hydrogen waa given up because of the too high sensitivity of the sample to air (equation 3). As soon as a sealed ampoule was cut, diethylamine which was recognized by the strong odor vaporized. Nitrogen was analyzed by titration according to equation 2: an exactly weighed sealed ampoule containing about 2.0 g. of the sample waa crushed with a glass rod in 30 ml. of standardized 0.5 N hydrochloric acid, and the excess acid was back-titrated with 0.5 N sodium carbonate with methyl orange as an indicator. Tin analyses were carried out according to Farnsworth and Pekola.9 In this case the vaporization of diethylamine has no effect on the analytical values of tin. Triethyl(diethy1amino)tin.-A tetrahydrofuran solution of diethylaminomagnesium bromide prepared from 2.9 g. (0.12 g.-atom) of magnesium, 10.9 g. (0.1 mole) of ethyl bromide, and 7.3 g. (0.1 mo!e) of diethylamine in 50 ml. of tetrahydrofuran was added dropwise to a solution of 14.5 g. (0.06 mole) of triethyltin chloride10 (b.p. 86-88’/9 mm.) dissolved in 50 ml. of tetrahydrofuran. An exothermic reaction was observed during this procedure. After refluxing for 4 hr. the solvent was distilled and the reaction product, was fractionated under reduced pressure. A fraction boiling a t 72-88’/11 mm. was collected; yield, 10.5 g. (6370). Redistillation gave 9.5 g. of an analytically pure fraction boiling a t 114-117’/23 mm. All procedures were carried out in a nitrogen atmosphere. This sample gave a negative Beilstein test for halogen. The product was stored in sealed ampoules under nitrogen. Anal. Calcd. for CloH~5KSn: Sn, 42.69; N, 5.04. Found: Sn,42.64,42.98,42.45; N,4.92. Tri-n-propyl(diethy1amino)tin and tri-n-butyl(diethy1amino)tin were also obtained in about 707, yields by an analogous way from tri-n-propyltin chloride10 (b.p. 122123”/10 mm.) and tri-n-butyltin chloride10 (b.p. 130-132’/4 mm. ), respectively. Tri-n-propyl(diethy1amino)tin had b .p. 11&120°/13 mm . Anal. Calcd. for C13HslNSn: Sn, 36.97; ?;, 4.36. Found: Sn, 37.41; N, 3.96. Tri-%-butyl(diethy1amino)tin distilled a t 124-1 34 / 8 mm. Anal. Calcd. for C M H U N S ~ Sn, : 32.77; N, 3.87. Found: Sn, 32.62; N, 3.40. O

(9) M. Farnsworth and J. Pekola, Anal. Chem., 31, 410 (19.59). (10) IC. -4.Korheskov, Ber., 66, 1661 (1930).

ERNEST I. BECKER3

Department of Chemistry, Polytechnic Institute of Brooklyn, Brooklyn 1, *Yew Y o r k Received M a y 31, 1962

It has been reported that allylamine4 (I) and benzylamine5 (11), n-butylamine and n-hexylamine6 are hydrogenolyzed to the parent hydrocarbons, propylene, toluene, butane and hexane, respectively, using triphenyltin hydride (111). Ammonia was claimed as one product of the reactions and hexaphenylditin (IV) mas also isolated. In attempting to repeat these results it was found that I or I1 with I11 does give IT7 and a steady evolution of gas. However, no propylene (using vapor phase chromatography), toluene (using vapor phase chromatography and infrared analysis), or ammonia (Kessler’s reagent) could be detected. In the reaction of I1 with 111,a considerable amount of benzene was formed. These results are in direct contrast to the successful reports of van der Kerk and Noltes, who, however, failed to identify positively the expected propylene and ammonia.’ These results are also in sharp contrast to those of Kupchik and Connolly. The latter reported that a superimposable infrared spectrum of toluene was obtained from the first drops of distillate of the reaction mixture, which is difficult to understand. We were unable t o find any trace of toluene. Benzene invariably accompanies the reaction of benzylamine with triphenyltin hydride and therefore should have been present in the first drops of distillate. Again, no positive evidence for ammonia was provided. In the course of our attempts to hydrogenolyze carbon-nitrogen bonds, we also examined ptoluidine, A’-methylbenzylamine, and AT,LV-dimethylbenzylamine. In these experiments IT’ and a gas were always obtained, but no hydrocarbon nor ammonia (see Experimental). In the case of p-toluidine and 111, which was carried out in a sealed tube, the gas was not condensable at -80’. Qualitatively, the rate of formation of I T (1) This work was supported b y O.O.R. Project No. 2440, Department of the Army Project No. DA-30-069-ORD-2851 and Grant DA-ORD-31-124-61-G39, U S. Army Research Office (Durham). (2) Taken from the dissertation submitted to the Faculty of the Polytechnic Institute of Brooklyn in partial fulfillment of the requirements of the degree of Doctor of Philosophy. ( 3 ) To whom inquiries should be sent. (4) J. G. Noltes and G. J. M. van der Keik, “Functionally Substituted Organotin Compounds,” T m Research Institute, Middlcsex, England, 1958, p. 115. (5) E. J. Kupchik and R. E. Connolly, J . Org. Chem., 26, 4747 (1961). (6) J. G. Noltes and G. J . 31. van der Kerk, Chem. Ind. (London), 294 (1959). (7) In a private communication from G. .I. 11.van der Kerk, 1 . C:. Noltes, and J. G , A . Luijten, the earlier hydrogenolysps are refuted.

decreased in the order of greater substitution of the amine group--i.e., 3" < 2' < 1". These results suggest that 111is being catalytically deconiposed t o I V and hydrogen.8 Ph3SnH

Ar-NH2

Ph3SnSnPhs

+ HP

The Photochemical Preparation of Dimethyl-N- (2-cyano-2-propyl)ketenimine from 2,2'-Azobisisobutyronitrile P. SMITH,J. E . SHEATS,~ A N D P. E. MILLER

Experimental Allylamine and 111.-To a round bottom flask equipped with a stirrer, reflux condenser, gas inlet tube, and gas exit tube leading from the top of the condenser to a trap containing 95y0 ethanol kept a t -78" were added 32.2 g. (0.092 mole) of triphenyltin hydride and 2.6 g. (0.046 mole) of allylamine. Evolution of a gas was observed on mixing. The reaction mixture was stirred under a nitrogen atmosphere til solidification was complete. The contents of the trap gave a negative test with Nessler's reagent for ammonia, no characteristic brown-red color being observed. The solid in the flask was recrystallized from benzene, m.p. 230232'. It did not depress the melting point on admixture with an authentic sample of 117. In a second experiment, using the same procedure as described above, 20 g. (0.057 mole) of I11 and 1.62 g. (0.029 mole) of I were used. Any possible propylene being evolved was led from the top of the condenser into a trap containing I-hexene kept a t -78", but no propylene was detected upon running the trap contents into a vapor phme chromatograph. Benzylamine and 111.-The procedure of Kupchik and Connolly was followed. The first few drops of distillate collected showed bands in the infrared (cm.-l) a t 3077, 3021, 1961, 1815, 1754, 1715, 1529, 1479, 1393, 1247, 1175, 1036, 850, 680, which excludes toluene'o and confirms benzene. Further distillation and subsequent infrared analyses still showed no evidence for toluene. In a second experiment, beginning with 22.5 g. (0.064 mole) of triphenvltin hydride and 3.6 ml. (0.032 mole) of I1 in which ammonia was particularly sought, a water-trap for the evolved gases contained no ammonia, (Sessler's reagent). p-Toluidine and 111.-Beginning with 2.15 g. (0.020 mole) of p-toluidine and 8.0 g. (0.023 mole) of 111, stirring a t 150" for 1 hr. and a t 190" for another 2 hr., throughout under a nitrogen atmosphere, evolution of gas was observed and solidification proceeded during the course of the reaction, but no positive test for ammonia was observed throughout this time. No toluene was determined in the distillate, only benzene. Similar results were obtained in a sealed tube experiment at 80-150 N-Methylbenzylamine and 111.-In two separate experiments, with 7.00 g. of I11 and 1.21 g. of Y-niethylbenzylamine each, no methylamine could be determined by means of the Rimini test. I t was noted that hydrogen was evolved only a t a higher temperature (-100') for the secondary amine. S,N-Dimethylbenzylamine and 111.-With this amine, using the procedure outlined in detail earlier, no test for secondary amine could be determined from 1.35 g. of dimethylbenzylamine and 7.0 g. of 111, by means of the Simon and the nickel dithiocarbonate tests. Again, it was noted that with the tertiary amine, decomposition of 111 took place a t a still higher temperature (-130-1.50"). O .

( 8 ) I n ref. 7 v a n der Kerk, Noltes, a n d Luijten give evidence for the stoichiometry. (9) We wish to express our gratitude to Drs. van der Kerk, Noltes, a n d Luijten for making their experimental work known to us prior to publication. (10) Comparisons were made with spectra a s follows: American Petroleum Institute Researah Project No. 44, Serial No. 308 (benzene) and No. 498 (toluene).

Depn~tmentof Chaniistr?l, Duke UniversitTl, Durham, North Carolina Receiaed June 65, 2962

Lately it has been necessary to have a preparative m e t h o ~ l -for ~ dimethyl-N-(2-cyano-2-propyl)ketenimine (I). Previously we hare employedjJ CHI

CHr

I

CHs-C--S=C=C-CHs

I

I

CN I

a procedure where 2,2'-azobisisobutyronitrile (11) CHI

CHr

I

CHs-C--N=N-C--CHa

I

CS

I

I CN

I1

was thermally decomposed in an inert solvent, cyclohexane, a t ca. 80" until the concentration of I had reached its maximum r a I ~ e , ~ ~ whereupon *--l~ the reaction mixture was quenched by cooling to ca. 10" and I isolated after filtering off the precipitate, which was largely all the tetramethylsuccinonitrile (111) formed and the rindecomposed 11. CHs CHs I

(

CHa--C--CCHr 1

cx

1

CK

IT1

The chief disadvantages of this method are that it requires care to keep gives low yields (ea. l n % ) , 5 9 7 the reaction under control, and demands a knowledge of the approximate time rvhen the maximum concentration of I is reached. (1) Supported b y grants from Esso Research a n d Engineering Company a n d the United States Public Health Service, Division of General Medical Sciences. (2) I n partial fulfilment of the requirements for Graduation with Distinction. (3) P. Smith. N. Muller, and W C . Tosch, J . Polymer Sei., 67, 823 (1962). (4) P. Smith a n d S.Carbone, J . A m . Chem. Soc., 81, 6174 (19.59). ( 5 ) P. Smith a n d A. M. Rosenberg. ibid., 81, 2037 (1959). (6) P. E. Miller, J. E. Munzenrider, J. E. Sheats, a n d P. Smith, unpublished work. (7) G. S. Hammond, 0. D. Trapp, R. T. Keys, and D. L. Neff, J. A m . Chem. Soc., 81, 4878 (1959). (8) G. 9. Hammond, C.-H. 9. Wu, 0. D. Trapp, J. Warkentin, a n d R. T. Keys, ibid.. 82, 5394 (1960). (9) M. Tallt-Erben and A. N. Isfendiyaroglu, Can. J. Chem., 36, 1156 (1958). (10) M. Talit-Erben and S. Bywater, J . Am. Chem. Soc., 7 7 , 3710, 3712 (1955).